Contribution of facility level factors to variation in antibiotic use in long-term care facilities: a national cohort study.

Abstract

To examine national variation in systemic antibiotic use in long-term care facilities (LTCFs) and identify facility characteristics associated with antibiotic utilization. This retrospective cohort study included 312 375 residents of 2536 Australian LTCFs between 2011 and 2016. LTCFs were categorized as low, medium or high antibiotic use facilities according to tertiles of DDDs of systemic antibiotics dispensed per 1000 resident-days. Multivariable logistic regression estimated the associations between facility characteristics (ownership, size, location, medication quality indicator performance, prevalence of after-hours medical practitioner services) and antibiotic use (low versus high). LTCFs in the lowest and highest antibiotic use categories received a median of 54.3 (IQR 46.5-60.5) and 106.1 (IQR 95.9-122.3) DDDs/1000 resident-days, respectively. Compared with not-for-profit LTCFs in major cities, government-owned non-metropolitan LTCFs were less likely to experience high antibiotic use [adjusted OR (aOR) 0.47, 95% CI 0.24-0.91]. LTCFs with 69-99 residents were less likely to experience high antibiotic use (aOR 0.69, 95% CI 0.49-0.97) than those with 25-47 residents annually. Greater prevalence of medical practitioner services accessed after-hours was associated with high antibiotic use [aOR 1.10 (per 10% increase in after-hours services), 95% CI 1.01-1.21]. South Australian LTCFs (aOR 2.17, 95% CI 1.38-3.39) were more likely, while Queensland (0.43, 95% CI 0.30-0.62) and Western Australian (aOR 0.34, 95% CI 0.21-0.57) LTCFs were less likely to experience high antibiotic use than New South Wales LTCFs. Considerable facility level variation in systemic antibiotic use was observed across Australian LTCFs. Identification of facility characteristics associated with antibiotic use provides a basis for targeted stewardship initiatives.